Forum Discussion
One-to-one wrong filter
Grieclolu Right, I used your 2 tables originally, not a single employee in Gender that matches with Terminations. Check it yourself in the updated PBIX attached below signature. Gender3, Table3, Page3.
Grieclolu Use the Filters pane and filter your legend to not include (Blank)?
It could't be a solution, since almost all the rows are red as blank.
In the screen below, i have 834 rows, that means 834 terminations. If i not include the blank, i will lose almost all the information (824). Do you need some example more? Or maybe the pbix file?
Grieclolu Right, I used your 2 tables originally, not a single employee in Gender that matches with Terminations. Check it yourself in the updated PBIX attached below signature. Gender3, Table3, Page3.
Grieclolu So you don't have those terminated employees in your Gender table? Sorry, confused. PBIX would definitely help. See mine attached under my signature.
Greg_Deckler I saw your attached, and my file is the same as yours. But it is wrong too. In fact, among the 15 employees present in the "table", you have 13 blanks, and only two correct values (1 male 1 female). I'd like to have all the 15 values filtered as male or female.
I don't know if you understood my request. Basically the result i want is just have all the values correctly filtered.Grieclolu I understand but if all of your terminated employees are in your "Gender" table then they would all be counted and if your terminated employees are NOT in your "Gender" table, then how in the world do you know their Gender to begin with? See the dilemma here?
Greg_Deckler I belive is correct that all the terminated are present in the gender table.
Let's make another example:
i have 4 emoloyees in Table 1, A, B, C, D; A and B are male, while C and D female.
then i have 2 employees in table 2 (terminations), A and D.My expectation is that, by creating the relationship, A and D could automaically be clustered as male (A) and female (B).
Can you do something like this?
Many thanks 🙂
Grieclolu So I implemented your scenario and it 100% works. See attached PBIX, Gender2, Table2, Page 2.
Greg_Deckler Perfect, in this example it works, so you see the result i want. Now, how can fix the previous example? Or any other bigger dataset?
Many thanks
Grieclolu See attached PBIX. I added user E to terminations list and you can see that since there are no matching records for E in Gender2 table, you get a blank in your donut chart. My suspicion is that you don't actually have all of the terminated employees in your Gender table or that you have trailing spaces or something in your data that is causing the records to be unmatched between the two tables.
Greg_Deckler thanks for your efforts and suggestions, but i don't thing the problem is linked to unmethed date. I double checked it, and you also have my two tables i shared with you (you also use in pbix), so you can chack by yourself.
Do you have any other suggestion?Thanks a lot, Luca